biosorption of mn (ii), co (ii) and cr (vi) in a ... · bioprocess model t. rezi ... (ii), co (ii)...

ISSN 0104-6632 Printed in Brazil

www.abeq.org.br/bjche

Vol. 31, No. 03, pp. 799 - 814, July - September, 2014 dx.doi.org/10.1590/0104-6632.20140313s00002905

*To whom correspondence should be addressed

Brazilian Journal of Chemical Engineering

BIOSORPTION OF Mn (II), Co (II) and Cr (VI) IN A HORIZONTAL ROTATING TUBULAR

BIOREACTOR: EXPERIMENTS AND EVALUATION OF THE INTEGRAL

BIOPROCESS MODEL

T. Rezić1, I. Rezić2, M. Zeiner3, S. Hann3, G. Stingeder3 and B. Šantek1*

1Department of Biochemical Engineering, Faculty of Food Technology and Biotechnology, University of Zagreb, Pierottijeva 6/IV, 10000 Zagreb, Croatia.

Phone: + 385 1 4605 290, Fax: + 385 1 4836 424 E-mail: [email protected]

2Department of Applied Chemistry, Faculty of Textile Technology, University of Zagreb, Prilaz Baruna Filipovića 28a, 10000 Zagreb, Croatia.

3Department of Chemistry, University of Natural Resources and Applied Life Sciences, BOKU - Vienna, Muthgasse 18, 1190 Wien, Austria.

(Submitted: August 8, 2013 ; Revised: October 20, 2013 ; Accepted: October 21, 2013)

Abstract - In this research, a multi heavy metals removal process with mixed microbial culture was examined in a horizontal rotating tubular bioreactor (HRTB) with different combinations of process parameters. Three metals were selected as examples of cations (manganese and cobalt) and oxy-anion (hexavalent chromium). Hydrodynamic conditions and biomass sorption capacity in the HRTB had the main impact on the heavy metals removal efficiencies, which were for Mn2+ 87.0-93.6%, Co2+ 89.0-95.7% and Cr6+ 99.7-100%, respectively. For the bioprocess description in the HRTB, the integral bioprocess model that combines hydrodynamics, mass transfer and kinetics was used. This model was evaluated for the new experimental conditions and average variances between experimental and simulated data were in the range of 0.12 - 3.21·10-3. The results obtained clearly show that the integral bioprocess model is able to describe the heavy metal removal process in the HRTB. Keywords: Heavy metals removal; Mixed microbial culture; Biosorption; Integral bioprocess model; Horizontal rotating tubular bioreactor.

INTRODUCTION

Wastewaters polluted with heavy metals have al-ways been a very serious problem due to the fact that these elements are not biodegradable and they can accumulate in living tissues, causing serious health problems. Many industrial processes are closely re-lated to environmental pollution with heavy metals. Therefore, there is a constantly growing concern

over the effect of heavy metals on humans and aquatic ecosystems. The Agency for Toxic Sub-stances and Disease Registry classifies heavy metals as human carcinogens and health risk substances (Valko et al., 2005; Deng et al., 2006).

There are many conventional methods (physical and chemical) for heavy metals removal, but in gen-eral they generate environmentally hazardous chemi-cal by-products that result in waste that is hard to

800 T. Rezić, I. Rezić, M. Zeiner, S. Hann, G. Stingeder and B. Šantek


treat. In addition, commonly used processes such as ion exchange or membrane technology are quite ex-pensive methods for large wastewater quantities that usually contain low concentrations of heavy metals (Lesmana et al., 2009; Wang and Chen, 2009; Gayathri and Senthil Kumar, 2010; Barakat, 2011).

As an alternative, different biological methods could be applied because they do not destroy metals, but concentrate and immobilize them (Gavrilescu, 2004; Vieira et al., 2012; Michalak et al., 2013; Sicupira et al., 2014; Vieira et al., 2014). Biosorp-tion is the removal of metals and their complexes by biological materials such as active or inactive micro-organisms, microbial aggregates or biofilms (Singh et al., 2006; Baysal, et al., 2009; Xie, et al., 2013). These materials can efficiently remove heavy metals from solutions and therefore they are ideal adsorp-tive media for wastewaters with low metal ion con-centrations. Microbial metal accumulation has re-ceived much attention during recent years, due to the potential use of microorganisms for treatment of metal-polluted water or wastewater streams (Gadd, 2008). Several bacterial species have been identified that remove toxic heavy metals (Singh et al., 2006; Yilmaz et al., 2010). For biosorption of heavy metals microbial biofilms show great potential in wastewa-ter treatment systems. Therefore, different bioreactor types (e.g., trickling filters, fluidized or packed bed bioreactors, thin layers or biodisc reactors) were de-veloped for different wastewater treatment systems (Gavril and Macoveanu, 2000; Nicolella et al., 2000; Chojnacka, 2010; Chu, 2010).

The horizontal rotating tubular bioreactor (HRTB) was designed as a combination of a thin layer (Moser, 1985) and a biodisc reactor (Kargi and Eker, 2001). Its interior is characterized by O-ring shaped partition walls that provide additional area for biofilm formation. Major advantages of the HRTB use in wastewater treatment are higher bio-process efficiency and stability compared to the bio-reactors with suspended microbial cells. Higher bio-process efficiency is a consequence of plug flow conditions in the HRTB that are related to the for-mation of concentration gradients along the bioreac-tor length and therefore inhibition and/or repression bioprocess kinetics can be avoided. Due to the biofilm formation on the inner surfaces of the HRTB biomass washout happens very rarely even when relatively high inflow rates are used. Biofilm thick-ness in the HRTB can be controlled by bioreactor rotation speed and consequently substrate and/or oxygen limitation in the biofilm can be prevented. Furthermore, tubular bioreactors (as well as the HRTB) are characterized by relatively low opera-

tional energy demands. Due to the great potential of the HRTB, different

bioprocesses were studied in this bioreactor such are anaerobic (Ivančić et al., 2004), aerobic (Slavica et al., 2004) as well as a combination of aerobic and anaerobic bioprocesses (Rezić et al., 2007). Fur-thermore, heavy metals removal (Fe2+, Cu2+, Ni2+ and Zn2+) with mixed microbial culture was also studied in the HRTB (Rezić et al., 2011a) and the integral bioprocess model was established that combines a mixing model and a metal ion diffusion-adsorption model (Rezić et al., 2011b). In recent years, many mathematical models have been established to repre-sent the biosorption of different metal ions in differ-ent bioreactor systems (Lesmana et al., 2009; Ong et al., 2013). The hydrodynamic behavior of these bio-reactor systems was mostly described by a dispersion model. For description of the biosorption process in these systems different equations can be used such are Bohart-Adams (Wang, 2007; Kargi and Cikla, 2007; Chu, 2010; Chatterje and Schiewer, 2011), Langmuir (Chen and da Silva et al., 2002; Kogej et al., 2010; Ghasemi et al., 2011), Freundlich (Kogej and Pavko, 2001; Quintales et al., 2009a; Rezić et al., 2011a,b), a pseudo-second-order rate expression (Martins et al., 2014), or Yoon-Nelson (Chatterje and Schiewer, 2011).

The novelty of this research is the study of the harmfulness of two less toxic metals (Co2+ and Mn2+) and one more toxic metal (Cr6+) from textile waste-water on the mixed microbial culture in the HRTB in order to check the bioprocess stability and repeatabil-ity. Furthermore, in this study for bioprocess moni-toring in the HRTB a more efficient and sensitive analytical method (ICP-MS) was used. Bioprocess performance in the HRTB was described by the inte-gral bioprocess model that combines mixing and dif-fusion-adsorption models (Rezić et al., 2011b) in order to confirm its ability to describe different heavy metals removal processes in the HRTB, as well as to prove its potential for the use in the scale-up procedure.

MATERIALS AND METHODS

Mixed microbial culture was isolated from ground soil samples located inside iron, vinyl and cement factories area (industrial area located near the town of Split, Croatia). Soil samples contain 4 - 35 times higher heavy metals concentrations than un-polluted soil (Lovrenčić et al., 2005). Therefore, it is expected that microbial cultures present in these soil samples are capable of growth in the medium with

Biosorption of Mn (II), Co (II) and Cr (VI) in a Horizontal Rotating Tubular Bioreactor: Experiments and Evaluation of the Integral Bioprocess Model 801

Brazilian Journal of Chemical Engineering Vol. 31, No. 03, pp. 799 - 814, July - September, 2014

increased concentrations of heavy metals (Kavamura and Esposito, 2010). Synthetic liquid medium was used for isolation of mixed microbial culture from soils samples that contained the following ingre-dients (g L-1): glucose, 10; FeSO4·7H2O, 0.62; CuSO4·5H2O, 0.49; NiSO4·6H2O, 0.56; ZnSO4·7H2O, 0.55; CoCl2·6H2O, 0.50; MnSO4·H2O, 0.39; K2Cr2O7, 0.20; Pb-acetate·3H2O, 0.23; yeast extract, 0.3; and tripton, 0.3 (pH = 4.40). The isolation procedure started by the addition of 5 g of soil samples into 200 mL of sterile synthetic medium and cultivation on the rotary shaker at 30 °C /48 h (rotation speed 120 min-1). After that, the obtained mixed microbial cul-ture was plated on the solid synthetic medium (syn-thetic liquid medium for isolation with 20 g L-1 of agar) at 30 °C /48 h. Dominant microbial species were selected and cultivated again several times (in liquid synthetic medium as well as plated on solid synthetic medium) in order to obtain the most active microbial species.

In this research, the isolated mixed microbial culture was cultivated at room temperature (20 ± 1 °C) in a synthetic medium containing (g L-1): glucose, 10; MnSO4·H2O, 0.39; CoCl2·6H2O, 0.51; K2Cr2O7, 0.20; yeast extract, 0.3; and tripton, 0.3 (pH = 4.2-4.6). In this investigation, media were sterilized at 121 °C for 20 minutes.

The HRTB was constructed as a 2.0 m long stainless steel tube with an inside diameter of 0.25 m. The interior of the HRTB was divided by O-ring shaped partition walls with an inside diameter of 0.19 m. The distance between the partition walls was 0.02 m and the liquid volume in the HRTB was 15 liters. The HRTB was horizontally placed on bearings that enable rotation of the whole bioreactor. The aera-tion was done through the central tube fixed on the axis of the HRTB. The aeration tube was submerged in the medium at five positions along the bioreactor to improve the aeration efficiency. During the entire research airflow rate was constant (152 L h-1). Sys-tems for sampling of broth and biofilm were placed along the bioreactor at 0.40 m intervals. The device for biofilm thickness determination was made of a flat plate (dimensions 0.02 x 0.02 m) fixed on the cover of the sampling place. The required amount of suspended microbial biomass (7.5 L) for the inocu-lation of the HRTB was obtained by batch cultiva-tion in the stirred tank bioreactor. After 24 hours the feeding process was started at a rate of 1 L h-1 and bioreactor rotation speed of 10 min-1. Mixed micro-bial culture initially grew in suspension and then a biofilm developed gradually on the bioreactor inner surfaces. It took 15 days (24 medium residence times) from the inoculation to obtain a stable biofilm in the

HRTB and then it was considered as ready to con-duct performance studies with different combina-tions of process parameters: medium inflow rate (0.5, 1.0 and 2.0 L h-1) and bioreactor rotation speed (5, 15 and 30 min-1). During this beginning period, the biofilm thickness measurement was also performed (every 12 hours) as well as monitoring the biofilm erosion process in order to evaluate biofilm stability in the HRTB. In this research, the inlet glucose concentration in the medium was constant (10 g L-1), as well as the inlet multi-metal ion concentrations (Co2+ = 0.125 g L-1, Mn2+ = 0.125 g L-1 and Cr6+ = 0.125 g L-1), respectively. The dynamics of the bio-process in the HRTB was monitored by withdrawing the samples along the bioreactor length after five residence times from when the new set of process parameters was established. In this research, for each combination of bioreactor process parameters (n and F), samples were withdrawn in triplicate (in total, at least 99 samples) at time intervals of every 12 hours after the new set of HRTB process parameters was established.

The biomass concentration in suspension was determined by centrifugation of 35 mL samples for 20 min at 4500 min-1 (3629 g), washing twice with demineralised water and then drying at 105 °C / 48 h. Supernatants were used for determination of Co2+, Mn2+ and Cr6+ concentrations by UV-VIS spectro-photometric methods (Fries and Getrost, 1977; Zeiner et al., 2010). All determinations were done in triplicate and analytical errors were in the range of 3.2 - 5.6%. Metal ion concentrations in the suspended biomass as well as in the microbial biofilm were de-termined with an inductively coupled plasma - mass spectrometer (ICP-MS). The instrument was equipped with a standard one piece extended torch with a quartz injector tube, a cyclone spray chamber and a GemCone nebulizer. The biofilm samples for ICP-MS were prepared in a closed microwave digestion system using a mixture of nitric acid (5 mL), hydrogen peroxide (1 mL) and doubly distilled water (1 mL). The samples were digested according to the following program (power [W]/time [min]): 250/3, 0/1, 250/4.5, 650/6, 400/5, ventilation 25.0 min and measured by ICP-MS.

Biofilm thickness measurements were done by a modified Venkataraman and Ramanujam (1998) method. The modifications were the use of graphite powder instead of chalk powder and microscope with a micrometric scale in place of the projector, respec-tively. The mass of biofilm was determined by col-lecting the biofilm sample from the inner bioreactor surface. The sample was suspended in demineralised water, centrifuged and washed twice with demineral-



ised water and then dried at 105 °C / 48 h. Biofilm density (cx,f) was calculated from the

measured dry weight of the attached microbial bio-mass by using the following equation:

f

fxfx LS

mc

⋅= ,

, (kg m-3) (1)

where mx,f - dry weight of microbial biomass in the biofilm, S - inside bioreactor surface and Lf - biofilm thickness.

Suspended biomass (qx,L) and biofilm (qx,f) sorp-tion capacity were calculated as follows:

Lx

MLx m

mq,

, = and fx

Mfx m

mq,

, = (mg g-1) (2)

where mM - mass of metal ion and mx,L - dry weight of suspended microbial biomass.

Volumetric removal rate of metal ions (QM) was estimated by Equation (3).

R

LMout

LMM t

ccQ ,,

0 −= (kg m-3 h-1) (3)

where 0

,LMc - metal ion concentration in the bioreac-

tor inflow, outLMc , - metal ion concentration in the bio-

reactor outflow and tR - retention time. For description of the HRTB performance, the

integral bioprocess model was used that combines mixing, mass transfer and bioprocess kinetics. The structured cascade “spiral flow” model was used for the mixing description in the HRTB (Šantek et al., 1996). Mass balances of metal ions for the first ideally mixed compartment of the first cascade (Figure 1a) were:

1,1,1,1 0 1, 2,1

, , ,

1,1 1,1 1,1 1,1, , ,( )

= + +

− + − −

L

L

M L Niu M L cr M L p M L

u p M L cr M L M L

dcV F c F c F c

dt

F F c F c V r (4)

where 1,1

LV - liquid volume in the first compartment

(Ni=1) of first cascade (Nl=1), 1,1,LMc - metal ion con-

centration in the first compartment (Ni=1) of the first cascade (Nl=1) in the liquid phase, Fu - bioreactor inflow, Fcr - circulation flow, Ni

LMc ,1, - metal ion con-

centration in the Ni-compartment of first cascade (Nl=1) in the liquid phase, Fp - back flow, 1,2

,LMc - metal ion concentration in the first compartment (Ni=1) of

the second cascade (Nl=2) in the liquid phase, 1,1,M Lr -

reaction rate in the first compartment (Ni=1) of first cascade (Nl=1) in the liquid phase and t - time.

First ideally mixed compartments of all cascades in the mixing model are assumed to be without a biofilm layer (Figure 1a). All other ideally mixed compartments inside the cascade are in direct contact with the biofilm layer (Figure 1b). Therefore, mass balances of metal ions are determined for the second (or other) ideally mixed compartments of the cascade (Figure 1a, b) as follows:

1,2,1,2 1,1 1,2

, ,

1,2 1,2 1,2 1,2 1,2, ,, ( 0)( )=

= −

− − −

M LL cr M L cr M L

m M L L M LM f Z

dcV F c F c

dt

S k c c V r (5)

where 1,2

LV - liquid volume in the second compart-

ment (Ni=2) of the first cascade (Nl=1), 2,1,LMc - metal

ion concentrations in the second compartment (Ni=2) of the first cascade (Nl=1) in the liquid phase, S1,2 - the surface for mass transfer in the second compart-ment (Ni=2) of the first cascade (Nl=1), km - mass transfer coefficient in the stagnant liquid layer,

2,1)0(, =ZfMc - metal ion concentration in the second com-

partment (Ni=2) of the first cascade (Nl=1) on the biofilm surface, 1,2

,M Lr - reaction rate in the second compartment (Ni=2) of the first cascade (Nl=1) in the liquid phase and t - time.

Metal ion concentrations in the bulk liquid phase (cM,L) of the compartment were assumed to be con-stant for each ideal mixed compartment (Figure 1a). In this model, it was also assumed that the biofilm is submerged in the stagnant liquid layer (Lg, Figure 1b). The mass transfer processes in the liquid phase and inside the microbial biofilm on the inner surfaces of the HRTB were defined as described earlier (Rezić et al., 2011b).

In this model, the substrate consumption rate in the biofilm ( 2,1

, fMr ) is assumed to be equal to the mass transfer of all dissolved metal ions in the biofilm:

2,1,2

2,1,

2

, fM

fM

Mef rz

cD =

∂

∂ (6)

where Def,M - effective diffusion coefficient of the metal ion in the biofilm, 2,1

, fMc - metal ion concentra-tions in the second compartment (Ni=2) of the first cascade (Nl=1) in the biofilm and z - biofilm depth coordinate.



1,Ni 1,1

1,2

1,3

Nl,NiNl,1

Nl,2

Nl,3

1,Ni-1

Nl,Ni-1

(b)

2,Ni 2,1

2,2

2,3 2,Ni-1

(a)

cs cM,f(z=0)

z=Lg z=0 z= Lf

cM,L1,2 cM,f(z=Lf)

z

Lg Lf

cM,f1,2

S1,2

VL1,2

Fu inflow Fp back flow Fcr circulation flow Fu + Fp Fu + Fp+ Fcr

1,Ni 1,1

1,2

1,3

Nl,NiNl,1

Nl,2

Nl,3

1,Ni-1

Nl,Ni-1

(b)

2,Ni 2,1

2,2

2,3 2,Ni-1

(a)

cs cM,f(z=0)

z=Lg z=0 z= Lf

cM,L1,2 cM,f(z=Lf)

z

Lg Lf

cM,f1,2

S1,2

VL1,2

Fu inflow Fp back flow Fcr circulation flow Fu + Fp Fu + Fp+ Fcr

Figure 1: “Spiral flow” mixing model of the HRTB (a) and metal ion concen-tration profile in the second (Ni=2; b) ideally mixed compartment of the first cascade (Nl=1) in the mixing model.

The outer boundary conditions (at z = 0) at the liquid - biofilm interface are given as:

1,2 1,2 1,2, , ( 0)

1,2,1,2

, 0

( )

( )

=

=

− =m M L M f z

M fef M z

S k c c

dc zS D

dz

(7)

The inner boundary conditions (at z = 1,2

fL ) at the biofilm-bioreactor wall (or discs) interface are given as:

2,1

)(0

2,1,

fLz

fM

dzzdc

== (8)

Metal ion concentrations in the biofilm (cM,f) are

presented as functions of biofilm depth z (Figure 1b) and are given as a second-order polynomial equation:

2210, )( zazaazc fM ++= (9)

where 0a , 1a and 2a are the second-order polyno-mial correlation coefficients.

For the compartments with biofilm layer in the state of dynamic equilibrium at time (t), the model was derived from mass balance equations (Equations (5), (6)- (8)) and the second-order polynomial corre-lation for description of the metal ion concentrations (Equation (9)) across the biofilm layer. For more details see Rezić et al. (2011b).

Mass balance equations of the integral bioprocess model were coupled to the reaction rates in the liquid phase (rM,L) and in the biofilm layer (rM,f), respec-tively. Instead kinetic terms for heavy metal removal were estimated by Freundlich adsorption isotherm (Equations (10) - (11)):

1/, , ,( ) Lh

x L F L M Lq K c= (10)



where qx,L - suspended biomass adsorption capacity, KF,L - Freundlich isotherm constant for suspended microbial biomass, cM,L - metal ion concentration in the liquid phase and hL - Freundlich isotherm con-stant for suspended microbial biomass.

1/, , ,( ) fh

x f F f M fq K c= (11) where qx,f - biofilm adsorption capacity, KF,f - Freundlich isotherm constant for microbial biofilm, cM,f - metal ion concentration in the biofilm and hf - Freundlich isotherm constant for microbial biofilm.

The kinetic model of the biofilm layer was de-rived from the Freundlich isotherm (Equation (11)) and the second-order polynomial correlation and it has following form:

[ ] fhiifFfx zazaaKq /12

210,, )(++= (12) zi is collocation point across the biofilm layer paral-lel to the substrate surface (zi = Lf /p where p=2).

The kinetic model assumes that the reaction rate is a function of biomass concentration in the liquid phase (cx,L) and in the biofilm layer (cx,f):

R

,,, t

qcr LxLx

LM = (13)

where rM,L - reaction rate in the liquid phase and tR - retention time.

R

,,, t

qcr fxfx

fM = (14)

where rM,f - reaction rate in the biofilm and tR - reten-tion time.

A system of differential equations was obtained by establishing mass balances for each ideally mixed compartment of the physical model (Figure 1a) and consequently the concentration of each metal ion along the HRTB can be determined.

Established model equations were solved by us-ing the software "Wolfram Mathematica". The or-thogonal collocation method was used for descrip-tion of the inner biofilm metal ion concentration pro-files (Arora et al., 2006). As orthogonal polynomial, the second-order polynomial equation was used. Af-ter preliminary determinations, optimization was per-formed by calculating the global minimum of vari-ance (En) between experimental and simulated vari-ables (using the software "Wolfram Mathematica") by the following equation:

∑=

=

−= uni

i in

isimn

in

un c

ccn

E1

exp,

2,exp, )(1

(15)

where nu - number of observations, i

nc exp, - experi-mental variables and i

simnc , - simulated variables.

RESULTS AND DISCUSSION

In this research, the dynamics of heavy metal (Co2+, Mn2+ and Cr6+) removal with mixed microbial culture in the HRTB was studied with different com-binations of bioreactor process parameters (n and F). Changes of bioreactor process parameters (n and F) during this investigation were in accordance with our previous research (Rezić et al., 2011a). In order to obtained microbial culture for this research, an isola-tion procedure was performed by using soil samples extremely polluted with heavy metals (see Materials and methods). The assumption was that microbial cultures present in these soil samples are capable to grow on the medium with relatively high concentra-tions of heavy metals (Kavamura and Esposito, 2010). Due to the relatively long research time in the HRTB, it was necessary additionally to protect the bioprocess performance by using a relatively low pH of the medium (4.2-4.6). During research in the HRTB, the morphological characteristics of the mi-crobial species were also monitored and it was ob-served that cocci and rods were the dominant mi-crobial species in the suspended biomass, as well as filamentous bacteria in the microbial biofilm, respectively.

The bioprocess performance in the HRTB was monitored by the metal ion removal efficiency. The average volumetric metal ion removal rate (QM) as a function of the inflow rate (F) and bioreactor rotation speed (n) are presented in Table 1. The volumetric metal ion removal rate (QM) varied from 3.363 to 8.324 mg L-1 h-1 at inflow rates up to 1.0 L h-1. The increase of the inflow rate to 2.0 L h-1 was related to the increase of the volumetric removal rate from 14.530 to 16.617 mg L-1 h-1 as a consequence of the higher metal ion volumetric load and the decrease of mass transfer resistance in these conditions (Kogej et al., 2010).

In this investigation, the metal ion removal effi-ciency was at approximately the same level for all metal ions examined [cobalt ion (89.0-95.7%), man-ganese (87.0-93.6%) and chromium (99.7-100%)] (Table 1). On the basis of these results it is clear that the microbial biomass (suspended biomass and



biofilm) in the HRTB has relatively high affinity for all three metal ions.

The increase of the inflow rate to 2.0 L h-1 was related to the approximately two-fold increase of the volumetric removal rate (6.67-16.6 mg L-1 h-1) as a consequence of the higher metal ion volumetric load. In this investigation, the highest volumetric produc-tivity of metal removal was observed for chromium (4.17-16.60 mg L-1 h-1) compared to cobalt (3.98-13.7 mg L-1 h-1) and manganese (3.38-14.99 mg L-1 h-1; Table 1). On the basis of these results, it is shown that the biofilm in the HRTB has the highest affinity for chromium ion removal. This phenomenon can be explained by transport limitations due to the ion molar volume, which is smaller for chromium than for cobalt and manganese. Furthermore, the covalent index of chromium and its diffusion rate through the microbial biofilm are higher compared to the other two metal ions examined (Kogej and Pavko, 2001). Table 1: Average volumetric removal rate (QM) and removal efficiency (EM) of metal ions at different combinations of bioreactor process parameters (n and F).

QM (mg L-1 h-1)

EM (%) F

(L h-1) n

(min-1) Co2+ Cr6+ Mn2+ Co2+ Cr6+ Mn2+

5 3.598 3.598 3.873 31.0 99.96 93.1 15 3.363 3.363 3.859 60.3 99.87 92.7 0.5 30 3.454 3.454 3.893 68.5 99.90 93.5 5 7.523 7.523 7.811 41.1 99.82 93.7 15 7.478 7.478 7.797 48.8 99.84 93.5 1.0 30 7.781 7.781 7.804 58.8 99.77 93.6 5 15.569 15.569 14.663 40.3 99.68 87.6 15 15.065 16.609 14.530 42.1 99.68 87.0 2.0 30 15.128 16.617 14.663 40.7 99.73 88.0

Similar investigation of nickel and chromium

biosorption was done by Barroso et al. (2006) in the UASB reactor with sanitary sewage sludge. Removal efficiency was in the range of our results. Samarth et al. (2012) investigated the biosorption of a multi-metal aqueous solution on the living biomass of Ba-cillus licheniformis which was already used in the sewage treatment process. Aerobic batch biosorption experiments were carried out for removal of Cr (VI), Fe (III), Cu (II) ions from aqueous multi-metal solu-tions. Removal efficiencies for iron, chromium and copper ion were 95, 52 and 32% after 48 hours of bioprocess, respectively. Gomathi et al. (2012) in-vestigated the biosorption of chromium by Aplano-chytrium sp. Maximum removal of chromium (69.4%) was observed in the batch system with the following bioprocess conditions: pH range of 6.8 - 8.2, T = 27 °C and adsorbent dosage (dried biomass

of Aplanochytrium sp.) of 0.51 g L-1. In the study of Zhou et al. (2007) the removal of Cr (VI) from aque-ous solution was studied in a batch system by the sorption on the dead cells of Bacillus licheniformis isolated from metal-polluted soils. The maximum biosorption capacity of Cr (VI) ion (60.5 mg/g) was observed under the following conditions: pH = 2.5, Cr (IV) concentration = 300 mg L-1, biomass con-centration = 1 g L-1, T = 50 °C and contact time = 2 h. In the study of Quintelas et al. (2009b) the following removal efficiencies were observed: iron (100%), cadmium (70%), nickel (40 - 74%) and chromium (20 - 100%) in the batch system by sorption on the biofilm of Escherichia coli supported on kaolin. Arthrobacter viscosus biofilm supported on granular activated carbon in a batch system was also exam-ined. Six isotherms were tested and the best descrip-tion of the experimental data was observed for the Freundlich isotherm. In this research, removal effi-ciency was reduced with an increase of the initial chromium concentration (95.20% for 5 mg L-1 and 38.28% for 1000 mg L-1). The batch adsorption studies were used to develop a pilot scale bioreactor which has the capacity for complete chromium re-moval from aqueous solutions (average chromium removal efficiency was 99.9% in the first 30 days of the bioprocess; Quintelas et al. 2009a). Three differ-ent species of nonliving red algal biomass Laurancia obtusa, Geldiella acerosa and Hypnea sp. were used in a fixed-bed column bioreactor for removal of toxic heavy metal ions [Cu (II), Zn (II), Mn (II) and Ni(II)] from industrial effluent. The highest efficiency of metal ion removal was detected in the bioreactor with L. obtusa (94%), followed by G. acerosa (85%) and Hypnea sp. (71%; Wael and Hawazin 2013).

Suspended biomass concentrations obtained for different combinations of bioreactor process pa-rameters (n and F) at the outlet of HRTB are pre-sented in Table 2. The suspended biomass concen-trations (cx.L) were in the range of 0.72-1.28 g L-1 (previous research 1.66-1.95 g L-1) at an inflow rate of 0.5 L h-1. Furthermore, the increase of the inflow rate to 1.0 and 2.0 L h-1 was related to the increase of suspended biomass concentrations to 1.10-6.32 g L-1 (previous research 2.05-3.48 g L-1) as a consequence of biofilm erosion (continuous release of smaller biofilm parts). The highest detected suspended bio-mass concentration in the HRTB was 6.32 g L-1 as a consequence of intensive biofilm detachment (re-lease of larger biofilm parts) due to experimental conditions (n = 30 min-1 and F = 2.0 L h-1) that con-siderably stimulated this process. In this case, a considerable increase of metal ion concentrations in the liquid phase was observed as a consequence of



biomass washout from the HRTB. On the basis of the obtained results it is clear that the bioreactor ro-tation speed did not have a pronounced effect on the suspended biomass concentration. These observa-tions are in agreement with our previous research (Ivančić et al., 2004; Rezić et al., 2011a). However, some discrepancies from the above mentioned effect were observed at F= 0.5 L h-1 and n = 30 min-1 as a consequence of analytical errors. Table 2: The suspended biomass concentrations (cx,L) and sorption capacity (qx.L) at different com-binations of bioreactor process parameters (n and F) during heavy metals removal process in the HRTB.

qx.L (mg g-1) F

(L h-1) n

(min-1) LHRCB (%)

cx,L (g L-1)

Co2+ Cr6+ Mn2+ 0.5 5 0 0.85 31.004 3.746 29.451 0.5 5 50 1.14 33.981 10.849 37.737 0.5 5 100 1.28 6.834 1.691 7.313 0.5 15 0 0.72 4.518 2.632 4.494 0.5 15 25 0.78 8.251 4.013 8.470 0.5 15 50 0.83 9.256 5.304 10.630 0.5 15 75 1.14 7.907 1.545 8.576 0.5 15 100 1.23 7.530 1.369 8.067 0.5 30 0 0.96 19.389 4.365 19.729 0.5 30 50 1.05 21.230 7.744 23.992 0.5 30 100 1.09 2.044 0.677 2.092 1.0 5 0 1.10 1.654 0.200 1.708 1.0 5 50 1.62 13.917 1.832 14.540 1.0 5 100 2.05 7.742 1.582 9.096 1.0 15 0 1.38 29.249 2.489 33.762 1.0 15 25 1.42 17.998 3.711 22.532 1.0 15 50 1.86 14.732 1.948 17.850 1.0 15 75 2.21 18.239 2.408 22.570 1.0 15 100 2.33 24.280 2.699 29.509 1.0 30 0 1.78 27.640 2.269 28.994 1.0 30 50 2.64 34.788 4.189 40.848 1.0 30 100 2.89 32.935 3.255 39.958 2.0 5 0 1.91 36.926 3.050 40.558 2.0 5 50 2.96 16.743 5.309 21.591 2.0 5 100 3.13 21.788 2.201 28.031 2.0 15 0 2.02 49.712 7.027 42.970 2.0 15 25 2.57 17.163 6.169 23.099 2.0 15 50 3.03 8.866 5.976 13.603 2.0 15 75 3.25 5.568 5.508 8.984 2.0 15 100 3.59 6.289 5.516 9.902 2.0 30 0 3.82 57.160 4.295 59.010 2.0 30 50 5.20 28.427 1.653 34.266 2.0 30 100 6.32 27.673 2.293 35.050

The suspended biomass sorption capacity (qx.L)

during this research is also presented in Table 2. The highest bioreactor rotation speed (30 min-1) de-creased the thickness of the stagnant liquid-layer at the biomass surface and facilitated conditions for

metal ion sorption. Higher biomass sorption capacity was observed at inflow rates of 1.0 and 2.0 L h-1 as a consequence of biofilm erosion. The content of the biofilm in the suspended biomass has a significant impact on the suspended biomass sorption capacity. Biofilm structure and the extracellular polymeric substance (EPS) content increase the possibility of ion sorption (Guibaud et al., 2009; Ozturk et al., 2009; Ni et al., 2010). EPS is polyanionic in nature, with enhanced capability for cation ion bonding. Moreover, enzymatic activities in EPS also assist in the detoxification of heavy metals by transformation and subsequent precipitation in the EPS (Pal and Paul, 2008).

The sorption capacity of suspended biomass (qx.L) was higher for cobalt and manganese ions than for chromium ion. This effect can be explained by the fact that cobalt and manganese ions have a slightly higher covalent index (Chen and Wang, 2007) as well as the pH gradient along the HRTB (Baes and Mesmer, 1976). In this investigation, at an inflow rate of 0.5 L h-1, the pH gradient along the HRTB was in the range of 4.43 - 4.72. Increase of inflow rate to 1 and 2 L h-1 was related to the increase of the pH gradient along the HRTB in the range of 4.19-5.82 (Table 3). Table 3: pH variations along HRTB at different combinations of bioreactor process parameters (n and F) during the heavy metal removal process.

F (L h-1)

n (min-1)

pH (0% LHRCB)

pH (50% LHRCB)

pH (100% LHRCB)

0.5 5 4.56 4.68 4.71 0.5 15 4.43 4.52 4.68 0.5 30 4.52 4.72 4.72 1.0 5 4.51 4.92 5.29 1.0 15 4.39 4.86 5.37 1.0 30 4.29 4.98 5.21 2.0 5 4.31 5.12 5.72 2.0 15 4.19 5.25 5.82 2.0 30 4.25 5.43 5.69

On the basis of these results it is clear that the

inflow rate has a more pronounced effect on the pH gradient than bioreactor rotation. Metal ion sorption on the microbial biomass is pH dependent, as the pH affects the availability of metal ions in solution (speciation), as well as the metal binding sites on the cell surface. As the pH is increased along the HRTB due to the medium consumption, the overall surface charge of microbial cell will become negative, resulting in more efficient cobalt and manganese sorption. In aqueous solution Cr6+ ion appears as an oxyanion (CrO4

2− or Cr2O72−) and therefore it can not

bind effectively to a negatively charged microbial



cell surface. The Cr6+ ion sorption process in the HRTB was mostly carried out only on the first 50% of bioreactor length where the pH was below 5 due to the fact that the overall microbial cell surface charge becomes more positive, promoting sorption of anionic Cr6+ ion species (Baes and Mesmer, 1976; Zhou et al., 2007). Furthermore, hydrodynamic con-ditions in the HRTB also affected all previously mentioned factors (Gavrilescu, 2004). Therefore, on the basis of these results it is clear that biological and hydrodynamic conditions in the HRTB have a sig-nificant effect on the suspended biomass concentra-tion and its sorption capacity (Table 2).

Biofilm sorption capacity (qx.f) and biofilm volu-metric density (cx.f) at the inlet and the outlet of the HRTB were determined at n = 30 min-1 and F = 2.0 L h-1 (Table 4). The HRTB is characterized by a concen-tration gradient along the bioreactor, so consequently a higher volumetric biofilm density was observed at the inlet of the HRTB (59.7±5.2 g L-1) than at the outlet of the HRTB (39.3±4.4 g L-1). This effect can be explained by the fact that higher substrate concen-trations for microorganism growth are present at the inlet than at the outlet of the HRTB. The structure and the content of the biofilm have a significant effect on the biofilm density and sorption capacity. Microbial sorption capacity is closely related to the content and molecular size of EPS in the biofilm. Due to these facts, the outer biofilm layers have higher porosity and provide relatively easy metal ion access to deeper biofilm layers reached with EPS. Therefore, high volumetric density biofilms have higher sorption capacity than low density biofilms characterized by the low content of EPS (Cloirec et al., 2003). Table 4: Biofilm sorption capacity (qx.f) and volumetric density (cx.f) at the inlet and outlet of the HRTB for F = 2.0 L h-1 and n = 30 min-1.

LHRCB (%) qx,f (mg g-1) 0 100 Co2+ 37.123 20.652 Cr6+ 14.512 5.649 Mn2+ 49.217 49.233

Σqx.f (mg g-1) 100.852 49.233 cx.f (g L-1) 59.7±5.2 39.3±4.4

At F = 2.0 L h-1 and n = 30 min-1 the lowest biofilm

sorption capacity was observed for chromium ion compared to the other metal ions examined (Table 4). As in the case of suspended biomass, this effect can be explained by the metal covalent index, the me-dium and biofilm characteristics and the hydrody-namic conditions in the HRTB (Kogej and Pavko,

2001; Cloirec et al., 2003; Chen and Wang, 2007; Chen and Wang, 2008). The main impact on the biofilm sorption capacity is from the covalent index of the metal ions and the biofilm volumetric density. Cobalt and manganese ions have slightly higher co-valent index than chromium ion and therefore metal ions with higher covalent index have higher potential to form bonds with biological ligands in the biofilm (Kogej and Pavko, 2001; Chen and Wang, 2007). The structure and the content of the biofilm are re-sponsible for biofilm volumetric density, which af-fects biofilm sorption capacity (Hall-Stoodley and Stoodley, 2002; Liu and Tay, 2002; Cloirec et al., 2003). Hydrodynamic conditions are closely related to the shear stress and abrasion process in the HRTB. Therefore, the outer biofilm layers are more sensitive to these two effects than the inner biofilm layers. Outer biofilm layers can be released even at rela-tively small shear stress (Donlan and Costerton, 2002; Liu and Tay, 2002).

Biofilm thickness along the HRTB obtained for different combinations of medium inflow rates (F = 0.5-2.0 L h-1) and bioreactor rotation speed (n = 5-30 min-1) are presented in Table 5. The biofilm thick-ness was in the range of 0.23-1.43 mm, which is thinner compared to the literature data for mixed culture biofilm, but thicker compared to mono-mi-crobial culture biofilms (Rezić et al., 2007). The bio-reactor rotation speed did not have a significant ef-fect on the biofilm thickness (Table 5). The effect of medium inflow rate (F) on the biofilm thickness was more obvious at 2.0 L h-1 where biofilm erosion is intensified, combined with the decrease of biofilm thickness. It is well known that lower medium inflow rates favor the formation of thicker biofilms with lower areal biofilm density. The increase of inflow rate produces thinner biofilm with higher areal den-sity (Gavril and Macoveanu, 2000; Donlan and Costerton, 2002; Liu and Tay, 2002). Table 5: Biofilm thickness changes (Lf) along the HRTB at different combinations of bioreactor process parameters (n and F) during the heavy metal removal process.

Lf (mm) F (L h-1)

n (min-1) (0% LHRCB) (50% LHRCB) (100% LHRCB)

5 0.75 1.08 0.89 0.5 15 0.89 0.73 0.81

30 0.93 1.29 0.85 5 0.85 1.34 0.95

1.0 15 0.92 1.43 0.84 30 0.86 1.21 0.91 5 0.23 0.37 0.28

2.0 15 0.35 0.28 0.25 30 0.38 0.37 0.35



The biofilm thickness in the HRTB (Lf) was mostly stabile at inflow rates of 0.5 and 1 L h-1, and only smaller biofilm parts were observed in the liquid phase as a consequence of the biofilm erosion process. The highest biofilm thickness was observed in the first 50% of bioreactor length as a conse-quence of substrate ability in this part of HRTB. In-crease of the inflow rate to 2 L h-1 was related to the intensification of the biofilm erosion process as a result of higher shear stress and metal load in these conditions that were related to the decrease of bio-film growth and thickness. On the basis of these results, it is clear that the increase of the inflow rate and metal ion concentration produces thinner bio-films with higher density.

One of the goals of this research was to evaluate the integral bioprocess model for heavy metals re-moval in the HRTB. The established integral biopro-cess model combines the mixing model and a metal ion diffusion-adsorption model. A diffusive biofilm concept was introduced in order to detect changes of metal ion concentration inside the microbial biofilm attached to the inner surfaces of the HRTB (Rezić et al., 2011b).

In this research, the integral bioprocess model was used for description of the manganese (Mn2+), cobalt (Co2+) and chromium (Cr6+) removal process in the HRTB. Due to the fact that experiments were performed by the same combinations of bioreactor process parameters (n and F) the same optimal val-ues of mixing model parameters (Nl, Ni, Fcr and Fp) were used in the simulation of bioprocess behavior in the HRTB (Table 6). For more details related to the mixing model parameters and their effect on the in-tegral bioprocess model prediction efficiency see Rezić et al., (2011b). Table 6: Optimal values of adjustable mixing model parameters for different combinations of bioreactor process parameters (n and F) utilized in the integral bioprocess model (Rezić et al., 2011b).

F (L h-1)

n (min-1)

Nl Ni Fcr (m3 h-1)

Fp (m3 h-1)

5 5 3 6.23·10-4 9.22·10-3 0.5 15 5 3 7.04·10-4 5.33·10-3

30 4 3 4.29·10-3 1.02·10-3 5 5 3 5.64·10-4 9.34·10-3

1.0 15 5 3 8.07·10-4 5.23·10-3 30 4 3 5.01·10-3 9.61·10-4 5 8 6 8.95·10-4 8.89·10-3

2.0 15 8 6 2.41·10-3 4.78·10-3 30 7 5 5.83·10-3 9.09·10-4

The effective diffusion coefficients of metal ions

in the biofilm (Def,M) were calculated by previously

established procedure and presented in Table 7 (Rezić et al., 2011b). Obtained effective diffusion coeffi-cients (Def,M) are in the range of literature data (Camur and Yazicigil, 2005). Table 7: Values of effective diffusion coefficients for different metal ions in the microbial biofilm on the inner surface of HRTB.

Parameter Value (m2 s-1) Def.Co 9.2·10-10 Def.Cr 11.2·10-10 Def.Mn 8.9·10-10

Mass transfer rate coefficients of metal ions in the

HRTB (km) were determined by a previously estab-lished procedure and presented in Table 8 (Rezić et al., 2011b). The increase of bioreactor rotation speed (n) reduces the thickness of the stagnant liquid layer near the microbial biofilm in the HRTB and enlarges the Reynolds rotation number, which contributes to more efficient transfer of metal ions through the stagnant liquid layer. Therefore, the increase of bio-reactor rotation speed (n) was related to the increase of the mass transfer rate coefficient (km). Table 8: Mass transfer rate coefficients calculated for different bioreactor rotation speeds.

n (min-1) km (m h-1) 5 0.356 10 0.411 15 0.498 20 0.535 30 0.628

For description of metal ion adsorption in the

suspended biomass, the Freundlich isotherm was selected because of its better description of biopro-cess performance in the HRTB (Rezić et al., 2011b). In this study, the adjustable parameters of Freundlich isotherm for all examined metal ions are estimated by the procedure described earlier (Rezić et al., 2011b) and presented in Table 9. Results obtained are in the range of our previous research (Rezić et al., 2011b).

The Freundlich isotherm was also used for de-scription of the metal ions adsorption process inside the microbial biofilm on the inner surface of the HRTB. In the integral bioprocess model it is as-sumed that inside the microbial biofilm attached on the inner surface of the HRTB concentration profiles have concave forms, in agreement with literature data (Wik et al., 2005; Ntwampe et al., 2008). Values of adjustable parameters of Freundlich isotherm for the adsorption process inside the microbial biofilm in the HRTB are also presented in Table 9.



Table 9: Parameters of the Freundlich adsorption model for suspended biomass and microbial biofilm on the inner surface of the HRTB for different metal ions in the multi-metals solution.

Parameter Suspended biomass Microbial biofilm hCo 1.202 1.352 KF.Co 0.949 1.245 hCr 1.578 1.742 KF.Cr 0.793 1.128 hMn 1.958 2.034 KF.Mn 1.343 1.772

Optimization of biofilm adsorption parameters

was carried out by using the minimum of variance between experimental and simulated values of metal ion liquid concentrations in the HRTB as a criterion. Relations between the alteration of adsorption model parameters (KF,f and hf) and the variance (En) were estimated by the second-order regression curve, as shown in Figure 2 (as the example Mn2+ ion was se-lected). For Mn2+ ion the minimum of the fitted curve was chosen as the optimum for KF,f and hf (Figure 2) values, respectively.

Figure 2: Relation between the alteration of the Freundlich model parameters (KF,f and hf) and the variance (En) for Mn2+ ion

A similar procedure was also used for all other metal ions and the results are presented in Table 9. Because of the bioprocess complexity, which is af-fected by different environmental conditions (e.g. medium composition and pH gradient along the bio-reactor, composition and physiological state of sus-pended biomass and biofilm along the bioreactor), the previously described procedure was established

in order to take into account all these effects on the bioprocess performance efficiency in the HRTB. For more details related to the adsorption model pa-rameters and their effect on the integral bioprocess model prediction efficiency see Rezić et al., (2011b).

Changes of metal ion concentrations in the HRTB were simulated for the entire operation period by using the established integral bioprocess model that combines mixing, mass transfer and bioprocess ki-netics. The global minimum of variance between experimental and simulated values of metal ions concentration in the HRTB was the criterion for the evaluation of the integral bioprocess model predic-tion efficiency. Obtained variance values for differ-ent combinations of the bioreactor process parame-ters are presented in Table 10. As can be seen in Table 10, variance values for different metal ions were in the following range: ECo = 0.12 - 1.41·10-3; ECr = 0.34 - 1.11·10-3; EMn = 0.71 - 3.21·10-3. Ob-tained results were in the range of our previous re-search (Rezić et al., 2011b). However, these results also pointed out that changes of both bioreactor process parameters (n and F) did not have a sig-nificant impact on the integral bioprocess model pre-diction efficiency. Table 10: Variance between experimental and simulated metal ion concentrations for different combinations of bioreactor process parameters (n and F) using the integral bioprocess model.

Co2+ ECo · 10-3 n (min-1) 5 15 30 F (L h-1)

0.5 0.12 0.87 0.16 1.0 0.28 1.30 0.32 2.0 0.32 1.41 0.72

Cr6+ ECr · 10-3 n (min-1) 5 15 30 F (L h-1)

0.5 0.34 0.44 0.79 1.0 0.56 0.77 0.66 2.0 1.89 1.11 0.87

Mn2+ EMn · 10-3 n (min-1) 5 15 30 F (L h-1)

0.5 0.71 1.05 1.44 1.0 0.75 3.21 0.97 2.0 0.80 1.98 0.80

Examples of metal ion profiles along the biore-

actor length in the experiment with different medium inflow rates (F = 0.5 - 2.0 L h-1) and constant biore-actor rotation speed (n = 15 min-1) are presented in Figure 3 (Co2+ concentration Figure 3a; Cr6+ concen-tration Figure 3b; Mn2+ concentration Figure 3c). Points in these figures represented experimental



values and simulated curves the bioprocess model values, respectively. The metal ion concentrations in the inflow of the HRTB were 0.125 g L-1. At the first measuring point (0% LHRCB) metal ion concentrations were considerably lower due to the dilution effect and different metal ion sorption capacities. HRTB is characterized by plug-flow conditions and therefore a gradient of metal ion concentration along the bio-reactor was observed (Figure 3; Šantek et al., 2000).

Figure 3: Simulations of Co2+ (a), Cr6+ (b) and Mn2+ (c) ion concentrations along the HRTB at different medium inflow rates F = 0.5 L h-1 (○, full line), F = 1.0 L h-1 (*, dashed line), F = 2.0 L h-1 (∆, dotted line) and constant bioreactor rotation speed (n = 15 min-1) using in the integral bioprocess model.

The highest metal ion concentrations were de-tected in the first 25% of the bioreactor length and they could considerably reduce biomass activity and consequently the metal ion removal efficiency. In this experiment, the impact of medium inflow rate on the Co2+ and Cr6+ concentration profiles was not statistically significant. However, this impact was

more pronounced at other two bioreactor rotation speeds (5 and 30 min-1). As mentioned earlier, the biofilm sorption is a complex process that is affected by hydrodynamic conditions, the morphological and physiological characteristics of the biofilm (Aksu et al., 1992; van Hullebusch et al., 2003) and the covalent index of the metal ions (Kogej and Pavko, 2001; Kogej et al., 2010; Chen and Wang, 2007). As can be seen in Figure 3, relatively good agreement between experimental and simulated values of the metal ion concentration along HRTB was observed. Similar agreement was also found at all other bioreactor rotation speeds (data not shown). On the basis of these results, it is clear that the established integral bioprocess model is capable to describe the heavy metal removal process in the HRTB.

CONCLUSIONS

Wastewater polluted with heavy metals is a serious ecological problem. In this research, a multi heavy metal removal process with mixed microbial culture was studied in the HRTB by different combinations of process parameters. Mixed microbial culture has the ability to remove heavy metals from wastewater polluted with multi-metal solution (total concentra-tions up to 375 mg L−1). For the description of heavy metal removal in the HRTB, the integral bioprocess model was used, which combines hydrodynamics (mixing), mass transfer and kinetics. Mixing in the HRTB was described by the structured cascade model and metal ion removal by a combined diffu-sion-adsorption model, respectively. This bioprocess model was tested in new experimental conditions and the average variances between experimental and simulated metal ion concentrations were in the range of 0.12 - 3.21·10-3. Obtained results clearly show that the established integral bioprocess model can de-scribe the heavy metal removal process in the HRTB. On the basis of these results it is clear that the integral bioprocess model has great potential to be used in a scale-up procedure based on the geo-metrical, hydrodynamic and physiological similarity of the bioprocess on the small and large scale.

ACKNOWLEDGMENT

The authors are grateful for financial support from the Ministry of Science, Education and Sport, Republic of Croatia (Grant No. 0058-0581990-2004) and bilateral project (Croatia/Austria) “Heavy metals removal from textile wastewaters in HRTB”.



NOMENCLATURE

0a , 1a and 2a

the second order polynomial correlation coefficient

2,1, fMc metal ion concentrations in

the second compartment (Ni=2) of the first cascade (Nl=1) in the biofilm

kg m-3

1,1,LMc metal ion concentrations in

the first compartment (Ni=1) of the first cascade (Nl=1) in the liquid phase

kg m-3

NiLMc ,1

, metal ion concentrations in the Ni-compartment of the first cascade (Nl=1) in the liquid phase

kg m-3

2,1,LMc metal ion concentration in

the second compartment (Ni=2) of the first cascade (Nl=1) in the liquid phase

kg m-3

1,2,LMc metal ion concentration in

the first compartment (Ni=1) of the second cascade (Nl=2) in the liquid phase

kg m-3

2,1)0(, =ZfMc metal ion concentration in

the second compartment (Ni=2) of the first cascade (Nl=1) on the biofilm surface

kg m-3

0,LMc metal ion concentration in

the bioreactor inflow kg m-3

outLMc , metal ion concentration in

the bioreactor outflow kg m-3

cM,f metal ion concentration in the biofilm

kg m-3

cM,L metal ion concentration in the liquid phase

kg m-3

inc exp, experimental variables kg m-3

isimnc , simulated variables kg m-3

cx,f biofilm density kg m-3

cx,L suspended biomass concentration

kg m-3

Def,M effective diffusion coefficient of the metal ion in the biofilm

m2 s-1

En variance between experimen-tal and simulated data

EM removal efficiency of metal ions

%

Fcr circulation flow m3 h-1

Fp back flow m3 h-1

Fu bioreactor inflow m3 h-1

hf Freundlich isotherm constant for microbial biofilm

hL Freundlich isotherm constantfor suspended microbial biomass

KF,f Freundlich isotherm constant for microbial biofilm

KF,L Freundlich isotherm constant for suspended microbial biomass

km mass transfer coefficient in the stagnant liquid layer

mh-1

Lf biofilm thickness m1,2fL biofilm thickness in the

second compartment (Ni=2) of the first cascade (Nl=1)

m

Lg stagnant liquid layer thickness

m

mM mass of metal ion kgmx,f dry weight of microbial

biomass in the biofilm g

mx,L dry weight of suspended microbial biomass

kg

n bioreactor rotation speed s-1

nu number of observations Ni number of ideally mixed

compartments in cascade Nl number of cascades Qm volumetric removal rate of

metal ion mg L-1 h-1

qx,f biofilm adsorption capacity mg g-1

qx,L suspended biomass adsorption capacity

mg g-1

p collocation point rM,f reaction rate in the biofilm kg m-3 h-1

rM,L reaction rate in the liquid phase

kg m-3 h-1

1,2,M fr reaction rate in the second

compartment (Ni=2) of the first cascade (Nl=1) in the biofilm

kg m-3 h-1

1,1,M Lr reaction rate in the first

compartment (Ni=1) of the first cascade (Nl=1) in the liquid phase

kg m-3 h-1

S inside bioreactor surface m2

S1,2 the surface for mass transfer in the second compartment (Ni=2) of the first cascade (Nl=1)

m2

SNl,Ni the surface for mass transfer in the ideally mixed compartment of the cascade

⋅S

Nl Ni; m2

t time stR retention time hVL liquid volume in the HRTB m3



1,1L

V liquid volume in the first compartment (Ni=1) of the first cascade (Nl=1)

m3

1,2L

V liquid volume in the second compartment (Ni=2) of the first cascade (Nl=1)

m3

,LNl NiV liquid volume in the ideally

mixed compartment of the cascade

⋅LV

Nl Ni; m3

z biofilm depth coordinate mzi collocation point across

biofilm zone parallel to the substrate surface

REFERENCES Aksu, Z., Sag, Y. and Kutsal, T., The biosorption of

copper(II) by C. vulgaris and Z. ramigera. Envi-ronmental Technology, 13, 579-586 (1992).

Arora, S., Dhaliwal, S. S. and Kukreja, V. K., Appli-cation of orthogonal collocation on finite ele-ments for solving non-linear boundary value problems. Applied Mathematics and Computa-tion, 18, 516-523 (2006).

Baes, C. F. and Mesmer, R. E., The Hydrolysis of Cations. John Wiley & Sons, New York (1976).

Camur, M. Z. and Yazicigil, H., Laboratory determi-nation of multicomponent effective diffusion co-efficients for heavy metals in a compacted clay. Turkish Journal Earth Sciences, 14, 91-103 (2005).

Barakat, M. A., New trends in removing heavy met-als from industrial wastewater. Arabian Journal of Chemistry, 4, 361-377 (2011).

Barros, A. J. M., Prasad, S., Leite, V. D. and Souza, A. G., The process of biosorption of heavy metals in bioreactors loaded with sanitary sewage sludge. Brazilian Journal of Chemical Engineer-ing, 23, 153-162 (2006).

Baysal, Z., Cinar, E., Bulut, Y., Alkan, H. and Dogru, M., Equilibrium and thermodynamic studies on biosorption of Pb(II) onto Candida albicans biomass. Journal of Hazardous Materials, B, 161, 62-67 (2009).

Chatterjee, A. and Schiewer, S., Biosorption of cad-mium (II) ions by citrus peels in a packed bed column: Effect of process parameters and com-parison of different breakthrough curve models. Clean-Soil, Air, Water, 39, 874-881 (2011).

Chen, C. and Wang, J., Influence of metal ionic characteristics on their biosorption capacity by Saccharomyces cerevisiae. Applied Microbiology and Biotechnology, 74, 911-917 (2007).

Chen, C. and Wang, J., Removal of Pb2+, Ag+, Cs+ and Sr2+ from aqueous solution by brewery's waste biomass. Journal of Hazardous Materials, 151, 65-70 (2008).

Chojnacka, K., Biosorption and bioaccumulation- the prospects for practical applications. Environment International, 36, 299-307 (2010).

Chu, K. H., Fixed bed sorption: setting the record straight on the Bohart-Adams and Thomas models. Journal of Hazardous Materials, 177, 1006-1012 (2010).

Cloirec, P., Andres, Y., Faur-Brasquet, C. and Gerente, C., Engineered biofilms for metal ion removal. Reviews in Environmental Science and Biotechnology, 2, 177-192 (2003).

Deng, L., Su, Y., Su, H., Wang, X. and Zhu, X., Bio-sorption of copper (II) and lead (II) from aqueous solutions by nonliving green algae Cladophora fasciculares: Equilibrium, kinetics and environ-mental effects. Adsorption, 12, 267-277 (2006).

Donlan, R. M. and Costerton, J. W., Biofilm: Sur-vival mechanisms of clinically relevant microor-ganisms. Clinical Microbiology Reviews, 15, 167-193 (2002).

Fries, J. and Getrost, H., Organic Reagents for Trace Analysis. Merck Laboratory Press, Darmstadt (1977).

Gadd, G. M., Biosorption: Critical review of scien-tific rationale, environmental importance and sig-nificance for pollution treatment. Journal of Chemical Technology and Biotechnology, 84, 13-28 (2008).

Gayathri, R. and Senthil Kumar, P., Recovery and reuse of hexavalent chromium from aqueous so-lutions by a hybrid technique of electrodialysis and ion exchange. Brazilian Journal of Chemical Engineering, 27, 71-78 (2010).

Gavril, M. and Macoveanu, M., Attached-growth process engineering in wastewater treatment. Bioprocess Engineering, 23, 95-106 (2000).

Gavrilescu, M., Removal of heavy metals from the environment by biosorption. Engineering in Life Sciences, 3, 219-232 (2004).

Ghasemi, M., Keshtkar, A. R., Dabbagh, R. and Jaber Sardari, R., Biosorption of uranium(VI) from aqueous solutions by Ca-pretreated Cysto-seira indica alga: Breakthrough curves studies and modeling. Journal of Hazardous Materials, 189, 141-149 (2011).

Gomathi, V., Saravanakumar, K. and Kathiresan, K., Biosorption of chromium by mangrove-derived Aplanochytrium sp. Africian Journal of Biotech-nology, 11, 16177-16186 (2012).



Guibaud, G., van Hullebusch, E., Bordas, F., d’Abzac, P. and Joussein, E., Sorption of Cd(II) and Pb(II) by exopolymeric substances (EPS) extracted from activated sludges and pure bacterial strains: Modeling of the metal/ligand ratio effect and role of the mineral fraction. Bioresource Technology, 100, 2959-2968 (2009).

Hall-Stoodley, L. and Stoodley, P., Developmental regulation of microbial biofilms. Current Opinion in Biotechnology, 13, 228-233 (2002).

Ivančić, M., Šantek, B., Novak, S. and Marić, V., Fermentative bioconversion in a horizontal rotat-ing tubular bioreactor. Process Biochemistry, 39, 995-1000 (2004).

Kargi, F. and Eker, S., Comparison of the perform-ance of rotating perforated tubes and rotating bio-disc biofilm reactors for wastewater treatment. Process Biochemistry, 37, 1201-1206 (2001).

Kargi, F. and Cikla, S., Determination of model pa-rameters for zinc (II) ion biosorption onto pow-dered waste sludge (PWS) in a fed-batch system. Journal of Environmental Management, 85, 883-890 (2007).

Kavamura, V. N. and Esposito, E., Biotechnological strategies applied to the decontamination of soils polluted with heavy metals. Biotechnology Ad-vances, 28, 61-69 (2010).

Kogej, A. and Pavko, A., Comparison of Rhizopus nigricans in a pelleted growth form with some other types of waste microbial biomass as biosor-bents for metal ions. World Journal of Microbiol-ogy and Biotechnology, 17, 677-685 (2001).

Kogej, A., Likozar, B. and Pavko, A., Lead biosorp-tion by self-immobilized Rhizopus nigricans pel-lets in an laboratory scale packed bed column: Mathematical model and experiment. Food Tech-nology and Biotechnology, 48, 344-351 (2010).

Lesmana, S. O., Novie, F., Soetaredjo, F. E., Sunarso, J., and Ismadji, S., Studies on potential applica-tions of biomass for the separation of heavy metals from water and wastewater – Review. Biochemical Engineering Journal, 44, 19-41 (2009).

Liu, Y. and Tay, J., The essential role of hydrody-namic shear force in the formation of biofilm and granular sludge. Water Research, 36, 1653-1665 (2002).

Lovrenčić, V., Oreščanin, D., Barišić, L., Mikelić, B., Rozmaric-Macefat, M., Lulić, S. and Pavlović, G., Characterization of tenorm and sediments of Kastela bay and the influence of tenorm on the quality of sediments. Global Nest Journal, 7, 188-196 (2005).

Martins, R. J. E., Vilar, V. J. P. and Boaventura, R. A. R., Kinetic modelling of cadmium and lead

removal by aquatic mosses. Brazilian Journal of Chemical Engineering, 31, 229-242 (2014).

Michalak, I., Chojnacka, K. and Krowiak, A. W., State of the art for the biosorption process - A re-view. Applied Biochemistry Biotechnology, 170, 1389-1416 (2013).

Moser, A., Imperfectly Mixed Bioreactor Systems. In: M. Moo- Young (Ed.), Comprehensive Bio-technology (vol. 2, p. 77-98), Pergamon Press, Oxford (1985).

Ni, S. Q., Lee, P. H., Fessehaie, A., Gao, B. Y. and Sung, S., Enrichment and biofilm formation of Anammox bacteria in a non-woven membrane re-actor. Bioresource Technology, 101, 1792-1799 (2010).

Nicolella, C., Loosdrecht, M. C. M. and Heijnen, J. J., Wastewater treatment with particulate biofilm reactor. Journal of Biotechnology, 80, 1-33 (2000).

Ntwampe, S. K. O., Sheldon, M. S. and Volschenk, H., Oxygen mass transfer for an immobilised biofilm of Phanerochaete chrysosporium in a membrane gradostat reactor. Brazilian Journal of Chemical Engineering, 25, 649-664 (2008).

Ong, S. A., Toorisaka, E., Hirata, M. and Hano, T., Comparative study on kinetic adsorption of Cu(II), Cd(II) and Ni(II) ions from aqueous solu-tions using activated sludge and dried sludge. Applied Water Science, 33, 21-325 (2013).

Ozturk, S., Aslim, B. and Suludere, Z., Evaluation of chromium (VI) removal behaviour by two iso-lates of Synechocystis sp. in terms of exopolysac-charide (EPS) production and monomer compo-sition. Bioresource Technology, 100, 5588-5593 (2009).

Pal, A. and Paul, A. K., Microbial extracellular poly-meric substances: Central elements in heavy metal bioremediation. Indian Journal of Microbi-ology, 48, 49-64 (2008).

Quintales, C., Fonseca, B., Silva, H., Figueiredo, H. and Tavares, T., Treatment of chromium (VI) solutions in a pilot-scale bioreactor through a biofilm of Arthrobacter viscosus supported on GAC. Bioresource Technology, 100, 220-226 (2009a).

Quintelas, C., Rocha, Z., Silva, B., Fonseca, B., Figueiredo, H. and Tavares, T., Removal of Cd(II), Cr(VI), Fe(III) and Ni(II) from aqueous solutions by an E. coli biofilm supported on kao-lin. Chemical Engineering Journal, 149, 319-324 (2009b).

Rezić, T, Šantek, B., Novak, S. and Marić, V., Het-erotrophic cultivation of Paracoccus denitrificans in a horizontal rotating tubular bioreactor. World



Journal of Microbiology and Biotechnology, 23, 987-996 (2007).

Rezić, T., Zeiner, M., Šantek, B. and Novak, S., Heavy metals removal in a horizontal rotating tu-bular bioreactor. Water Air & Soil Pollution, 214, 343-355 (2011a).

Rezić, T., Zeiner, M., Šantek, B. and Novak, S., Mathematical modeling of Fe(II), Cu (II), Ni(II) and Zn(II) in a horizontal rotating tubular biore-actor. Bioprocess and Biosystem Engineering, 34, 1067-1080 (2011b).

Samarth, D. P., Chandekar, J. C., Bhadekar, R. K., Biosorption of heavy metals from aqueous solu-tion using Bacillus Licheniformis. International Journal of Pure and Applied Sciences and Tech-nology, 10, 12-19, (2012).

Sicupira, D. C., Tolentino Silva, T., Leão, V. A., Mansur, M. B., Batch removal of manganese from acid mine drainage using bone char. Brazilian Journal of Chemical Engineering, 31, 195-204 (2014).

da Silva, E. A., Cossich, E. S., Tavares, C. R. G., Filho, L. C. and Guirardello, R., Modelling of copper(II) biosorption by marine alga Sargassum sp. in fixed bed column. Process Biochemistry, 38, 791-799 (2002).

Singh, R., Paul, D. and Jain, R. K., Biofilms: Impli-cations in bioremediation. Trends in Microbiol-ogy, 14, 387-397 (2006).

Slavica, A., Šantek, B., Novak, S. and Marić, V., Microbial acetate oxidation in horizontal rotating tubular bioreactor. Journal of Biosciences, 29, 169-177 (2004).

Šantek, B., Horvat, P., Novak, S., Mayr, B., Moser, A. and Marić, V., Mathematical modeling of mixing in horizontal rotating tubular bioreactor: “Spiral flow” model. Bioprocess Engineering, 14, 223-229 (1996).

Šantek, B., Horvat, P., Novak, S., Moser, A. and Marić, V., Estimation of axial dispersion in hori-zontal rotating tubular bioreactor by means of structured model. Bioprocess Engineering, 23, 265-274 (2000).

Valko, M., Morris, H. and Cronin, M. T. D., Metals, toxicity and oxidative stress. Current Medicinal Chemistry, 12, 1161-1208 (2005).

van Hullebusch, E. D., Zandvoort, M. H. and Lens, P. N. L., Metal immobilization by biofilms: Mecha-

nisms and analytical tools. Reviews in Environ-mental Science and Biotechnology, 2, 9-33 (2003).

Venkataraman, R. and Ramanujam, T. K., A study on microbiology of biological film layer in rotat-ing biological contactors. Bioprocess Engineer-ing, 18, 181-186 (1998).

Vieira, A. P., Santana, S. A. A., Bezerra, C. W. B., Silva, H. A. S., Santos, K. C. A., Melo, J. C. P. and Airoldi, C., High performance maleated ligno-cellulose epicarp fibers for copper ion re-moval. Brazilian Journal of Chemical Engineer-ing, 31, 183-193 (2014).

Vieira, M. G. A., de Almeida Neto, A. F., Carlos da Silva, M. G., Nóbrega, C. C., Melo Filho, A. A., Characterization and use of in natura and calcined rice husks for biosorption of heavy metals ions from aqueous effluents. Brazilian Journal of Chemical Engineering, 29, 619-634 (2012).

Wang, J. and Chen, C., Biosorbents for heavy metals removal and their future. Biotechnology Ad-vances, 27, 195-226 (2009).

Wik, T., Goransson, E. and Breitholtz, C., Low model order approximations of continuously stirred biofilm reactors with Monod kinetics. Bio-chemical Engineering Journal, 30, 16-25 (2005).

Wael, M. I. and Hawazin, H. M., Bioremoval of heavy metals from industrial effluent by fixed-bed column of red macroalgae. Toxicology and Industrial Health, 29, 38-45 (2013).

Xie, P., Hao, X., Mohamad, O. A., Liang, J. and Wei, G., Comparative study of chromium bio-sorption by Mesorhizobium amorphae strain CCNWGS0123 in single and binary mixtures. Applied Biochemistry and Biotechnology, 169, 570-587 (2013).

Yilmaz, M., Tay, T., Kivanc, M. and Turk, H., Re-moval of copper(II) Ions from aqueous solution by a lactic acid bacterium. Brazilian Journal of Chemical Engineering, 27, 309-314 (2010).

Zeiner, M., Rezić, T. and Šantek, B., Monitoring of Cu, Fe, Ni, and Zn in wastewater during treat-ment in a horizontal rotating tubular bioreactor. Water Environment Research, 2, 83-186 (2010).

Zhou, M., Liu, Y., Zeng, G., Li, X., Xu, W. and Fan, T., Kinetic and equilibrium studies of Cr(VI) bio-sorption by dead Bacillus licheniformis biomass. World Journal of Microbiology and Biotech-nology, 23, 43-48 (2007).

biosorption of mn (ii), co (ii) and cr (vi) in a ... · bioprocess model t. rezi ... (ii), co (ii)...

Documents